The purpose of this study is to compare the perinatal outcome of a group of pregnant patients who are diagnosed with gestational diabetes using plasma insulin levels with a group of matched controls in whom the diagnosis is made by standard screening methods. Participants would receive 100 grams of glucose in the form of glucola and would have fasting, 1 hour, 2 hour, and 3 hour plasma insulin and glucose levels drawn. Four patterns of plasma insulin and glucose levels would be expected.